With the recognition of the role of antithrombotic agents in clinical medicine, investigators have pursued their efficacy, optimal dose, route of administration and safety. Aspirin has been found to be an effective antithrombotic agent in patients with cerebrovascular disease and ischemic heart disease. Aspirin may also have other antithrombotic applications. Although aspirin has become widely used as an antithrombotic agent, it still exhibits undesirable side effects, including gastrointestinal toxicity which is probably dose related.
To induce its suppressive effects, aspirin irreversible acetylates the enzyme cyclooxygenase found in platelets and vascular wall cells [Burch et al., J. Clin. Invest. 61:314 (1978); Majerus, J. Clin. Invest. 72:1521 (1983); Roth et al., J. Clin. Invest. 56:624 (1975)]. Cyclo-oxygenase converts arachidonic acid to thromboxane-A.sub.2 (TXA.sub.2) in platelets and to prostaglandin-I.sub.2 (PGI.sub.2 or prostacyclin) in vascular walls [see for example, FitzGerald et al., J. Clin Invest. 71:676 (1983); Preston et al., N. Engl. J. Med. 304:76 (1981)]. TXA.sub.2 induces platelet aggregation and vasoconstriction, while PGI.sub.2 inhibits platelet aggregation and induces vasodilation. In other words, aspirin can have both an antithrombotic effect (by reducing TXA.sub.2 production) and a thrombogenic effect (by reducing PGI.sub.2 production). As a result, striking an appropriate balance between aspirin's effects on TXA.sub.2 and PGI.sub.2 production has been a goal of aspirin therapy under these circumstances.
It is generally accepted that when aspirin is administered in doses of approximately 1,000 mg/day, it inhibits both TXA.sub.2 and PGI.sub.2 synthesis [Weksler et al., N. Engl. J. Med. 308:800 (1983)]. Daily administration of very low doses of aspirin (approximately 40 mg/day) has been reported to inhibit thromboxane-B.sub.2 synthesis in vitro and to reduce the urinary excretion of 2,3-dinor-thromboxane-B.sub.2 (both of which are metabolites of TXA.sub.2), without producing significant changes in the urinary excretion of 6-keto-prostaglandin-F.sub.1 a and 2,3-dinor-6-keto prostaglandin-F.sub.1 a (which are both metabolites of PGI.sub.2 production) [Patrignani et al., J. Clin. Invest. 69-1366 (1982); FitzGerald et al., supra]. While 40 mg/day has no significant effect on prostacyclin biosynthesis, it does have some measurable effect [FitzGerald et al., supra]. Moreover this dose does not suppress 2,3-dinor-TXB.sub.2 very well and it is not known whether it suppresses bradykinin-stimulated prostacyclin formation. Therefore, this dose has not been demonstrated to provide selective inhibition of thromboxane synthesis without also inhibiting prostacyclin formation.
In contrast, others have reported that equally low doses of aspirin reduced PGI.sub.2 synthesis by 50% in both arterial and venous tissue [Preston et al., supra], and even lower doses (20 mg/day for 1 week) have been reported to inhibit PGI.sub.2 synthesis in both arterial and venous tissue by 50% in atherosclerotic patients [Weksler et al., supra]. It has been proposed that although this differential effect on the inhibition of TXA.sub.2 and PGI.sub.2 synthesis has been reported when urinary metabolites are measured to assess inhibition, there is no significant evidence for this differential effect when PGI.sub.2 synthesis is measured by assay of vascular wall biopsy tissue or when the assays for TXA.sub.2 and PGI.sub.2 are performed on blood samples [Weksler et al., supra]. However, it is not possible to achieve platelet selectivity with standard oral aspirin. Inhibition of basal PGI.sub.2 biosynthesis is similar over doses of 80-2,400 mg/day and bradykinin-stimulated PGI.sub.2 formation is abolished on oral aspirin 75 mg/day.
Aspirin has also been found to be an effective treatment for other medical conditions which benefit from lowering of TXA.sub.2 levels. For example, it has been reported that daily doses of aspirin given during the third trimester of pregnancy can significantly reduce the incidence of pregnancy-induced hypertension and preeclamptic toxemia in women at high risk for these disorders as a result of reductions in TXA.sub.2 levels [Schiff et al., N. Engl. J. Med. 321:351 (1989)]. Aspirin has also been reported to provide positive effects in women at risk for pregnancy-induced hypertension. Low doses of aspirin were reported to selectively suppress maternal thromboxane levels, but only partially suppressed neonatal thromboxane, allowing hemostatic competence in the fetus and newborn [Benigni et al., N. Engl. J. Med. 321:357 (1989)]. The use of aspirin for reducing the risk of fatal colon cancer has also been proposed [Thun et al., N. Engl. J. Med. 325:1593 (1991)]. Reduction of thromboxane levels has also been suggested as a means for treating thrombosis in patients having antiphospholipid syndrome associated with lupus [Lellouche et al., Blood 78:2894 (1991)]. Low-dose aspirin has also been suggested as therapy for migraine headache [see for example, Buring et al., JAMA 264(13) (1990)]. The role of arachidonic acid metabolites (e.g., TXA.sub.2 PGI.sub.2) in migraine have also been investigated [see for example, Parantainen et al., "Prostaglandins in the Pathophysiology of Migraine" in P. B. Curtis--Prior (ed), Prostaglandins: Biology and Chemistry of Prostaglandins and Related Eicosanoids (New York: Churchill Livingstone 1988), pp. 386-401; Puig-Parellada et al., Headache 31(3):156 (1991); Tuca et al., Headache 29(8);498 (1989); Nattero et al, Headache 29(4):233 (1989)].
The use of aspirin as a thromboxane suppressant has been hampered by its tendency to cause gastric bleeding upon traditional administration of aspirin in oral dose form. Studies have reported that aspirin produces erythema of the gastric mucosa in approximately 80% of patients with rheumatic diseases, gastric erosions in approximately 40%, and gastric ulcer in 15% [Silvoso et al., Ann. Intern. Med. 91:517 (1979)]. Aspirin applied topically to gastrointestinal tissue damages gastric mucosa and induces occult gastrointestinal bleeding [Croft et al., Br. Med. J. 1:137 (1967)]. Intravenous administration of aspirin may also produce some effects on gastric mucosa which is less pronounced with parenteral than with oral administration [Grossman et al., 40:383 (1961)]. Oral administration of diluted solutions of aspirin cause considerably less bleeding than similar doses in tablet form, and aspiring solutions containing antacids with sufficient buffering capacity cause no measurable blood loss [Leonards et al., Arch. Intern. Med. 129:457 (1972)]. Enteric-coated aspirin use results in less gastric and duodenal mucosal injury than regular aspirin [Graham et al., Ann. Intern. Med. 104:390 (1986)].
It would, therefore, be desirable to provide an appropriate dosage form of aspirin which will provide thromboxane suppressing effects, preferably selective thromboxane suppressing effects, and will also avoid the adverse side effects observed with aspirin dosage forms currently employed in aspirin therapies.
Several reports have been made of the incorporation of aspirin into a various analgesic preparations. U.S. Pat. No. 4,948,588 discloses the use of ether derivatives of glycerols or polyglycerols as percutaneous absorption accelerators. Analgesics, such as morphine, codeine and aspirin, are suggested as possible active agents for use with these accelerators. An example disclosed incorporation of aspirin into a suppository which was administered to male rabbits.
U.S. Pat. No. 4,654,209 discloses creams containing nitroglycerine and other active ingredients. Analgesics, such as aspirin, are suggested as active ingredients. An example makes a cream containing 5-15% aspirin by weight which was applied to the skin of the abdomen, thigh or back of subjects, resulting in positive blood and urine tests for the active ingredient.
U.S. Pat. No. 4,476,115 discloses analgesic compositions applied to skin together with or subsequent to the application of a non-toxic water-soluble sulfite compound. Examples described the preparation of mixtures of aspirin and anhydrous sodium sulfite which was applied to the skin of a mammal and covered with a water impervious plastic sheet held in place by adhesive tape. Bioavailability was observed within 30 to 40 minutes as evidence by increased mobility of the subject and reduction of stiffness. Although such aspirin preparations have been used for their analgesic effects, such preparations have not to applicant's knowledge been applied for therapy in which thromboxane suppression is desired.